Design flows and security tradeoffs in the lower Mekong basin

Lotic environments support the food security of hundreds of millions globally, yet tradeoffs among freshwater-dependent food systems remain poorly understood. In the Lower Mekong Basin, where rice and fish production systems are highly heterogeneous, we modeled harvest outcomes using multivariate autoregressive state-space (MARSS) models and flood magnitude as a key driver, measured by the High Seasonal Amplitude Metric (HSAM). High HSAM values (0.5) positively affected floodplain (FP) fish catch and Ca` Mau rice harvests, while moderate val- ues (0.1) had negative effects. In contrast, Dai fish catch and Cambodian rice harvest responded positively to moderate HSAM values but negatively to high values. Based on these patterns, we engineered four flow regimes optimized for each system. Forecasts over 10 years showed that each engineered hydrograph increased harvest for its target system. Some tradeoffs emerged: the FP hydrograph boosted FP fish catch and Ca` Mau rice but reduced Dai fish and Cambodian rice; the Cambodian rice hydrograph showed the reverse. Alternating between high and mod- erate HSAM values mitigated risk to individual systems, improving FP fish catch while having mixed or neutral effects elsewhere. Setting HSAM at 0.19 stabilized production across all four systems, balancing tradeoffs and maintaining current yields. These results highlight the potential to deliberately manage hydrologic regimes to co-optimize food production systems. Expanding hydrologic objectives beyond power generation is essential for sustaining ecosystem services, maintaining regional food security, and staying within planetary boundaries. • Lotic environments provide food security to millions worldwide • Productivity of freshwater ecosystems is globally threatened • MARSS modeling results show promise for resource co-optimization • Science advised freshwater resource management creates sustainable futures